Keith,
This is NOT the screen used in Confirmatory Analysis.com. That screen has been used by my friends for quite some time privately.
Could you also further explain GARP (Growth At a Reasonable Price).
GARP is pretty much what the acroynm says. It is buying growth (and the ensuing price appreciation) at a reasonable price. If you focus on the earnings growth of a company as the key to unlocking its potential price appreciation, then what you want to do is to buy the stock when it is cheap relative to that potential and sell when it becomes overpriced vs. that potential.
Let's take an extreme example.
Company X which is growing 30% per year and will continue that trend for as far as the eye can see. In this perfect universe there are no external forces and company X gets no better or worse and always turns in their 30% growth rate. In this perfect universe, you also have company Y which is growing 10% per year and it also has the amazing ability to continue this growth year after year far into the future.
Now on the day that you decide to invest, both of these stocks are selling at their growth rate times next years earnings of $1. Which one do you buy? Run the numbers out 5 years and assume the forward p/e ratio rises toward the growth rate of the company over the next 5 years.
Company X
12 month forward eps = $3.71
forward p/e ratio = 30
paid = $1 * 30 = $30
stock price = $3.71 * 30 = $111.30
gain = ($111.30-$30)/$30 = 271%
Company Y
12 month forward eps = $1.61
forward p/e ratio = 10
paid = $1 * 10 = $10
stock price = $1.61* 10 = $16.1
gain = ($16.1-$10)/$10 = 61%
Kind of a no-brainer, just buy the company with the higher growth rate. But look at what happens when you buy at a 40% discount to eps growth (extreme GARP) on the forward p/e as I screen for.
Company X
12 month forward eps = $3.71
forward p/e ratio = 30
stock price = $3.71 * 30 = $111.30
paid = $1 * 30 / 1.4 = $21.42
gain = ($111.30-$21.42)/$21.42 = 420%
Nice difference 420% vs. 271%. Buying at a discount give you plenty of room for mistakes and that discount cushion is an important part of the formula I use.
What if you got all excited and bought the stock at $70 or more than twice the growth rate.
Company X
12 month forward eps = $3.71
forward p/e ratio = 30
paid = $70
stock price = $3.71 * 30 = $111.30
gain = ($111.30-$70)/$70 = 59%
Overpaying for growth (not GARP) leads you right back to the same sort of outcome as if you would have picked the worst outcome in the first part of the example.
What you pay for a stock does matter in your total outcome down the road contrary to what you would be led to believe with the current market.
When you say they are cheap enough, are you referring to 3, 12 or 36 months ago as to the price of the stock?
None of the above. You are looking backwards again <grin>. This screen only looks forward. "Cheap enough" is by definition buying the company at a discount to its long term growth potential using the basis of its forward p/e. I use a 40% discount in the calculation to find more extreme cases.
"We all know some analysts are good and some are bad." Is there a rating somewhere? I assume that we are talking about the analysts for the Technology Based Industry. Can we put the Good, Bad & the Ugly in our 101 handbook for Screening for Winners?
The WSJ sometime selects the best analysts in any given sector (although this seems more like a popularity contest) and sometimes I've run across a page which will say that analyst X has only given 3 estimates on company X and is therefore not rated, but practically when you are sorting through 200 companies or so that make the screen I would not bother to look at the good bad and ugly until I start to get down to the true due diligence. If you want a quick and dirty peek to see if you have a variety of analysts, look at the spreads in the analysts estimates. If the spreads are high, then you have a difference of opinion on the company. Someone will eventually be right and some will be wrong, or they will all be wrong ala CTAP this week.
I read reports from ML and BBRS and every once in a while I have to scratch my head and think "what is this fool talking about". Tom Kurlak at ML was considered one of the best, but frankly he would have lost you a lot of money over the last two years or so. I remember one conference call where Kurlak was on and he effectively said "chips are chips" equating PC chips are equivalent to communications chips. He equated the slowing of the PC market in early 1998 with the eventual slowing of the communications chip market. What a stupid call (can you say internet???), yet it did pressure the industry since ML is a very influential brokerage in the retail area.
I obtained a copy of an ML report on ORB (Orbital Sciences Corp) dated 3/18/99 which is a company that you previously stated you own. In this report they have cut eps for 1999 from $1.39 to $1.05.
A few subtle points first of all.
Mathematically, one analyst out of the 9 that follow the company doesn't move the aggregate value of the average estimate which the screen is based. In other words if the other 8 keep their estimates at $1.39, the one analyst moving on the stock doesn't change the outlook very much that may be the one "ugly" analyst that you were trying to weed out ahead of time up above, or the other 8 could be the bozos. When you run a screen you don't know. In this particular case perhaps others are cutting estimates also. I see from the data on Yahoo! that in the past 7 days the 1999 estimate has decreased from $1.46 to $1.39. Right now I don't know if the new ML number is in the $1.39 estimate.
Again mathematically, when you are looking forward 12 months, you have to include some of the year 2000 estimate now that part of the year 1999 is past. Since analysts raised the 2000 estimate in the past 7 days that is a partial offset to the decrease you might see in the 1999 estimates.
Does this in your opinion mean that ORB will move sideways for the next 3 months?
I wouldn't be surprised.
Which brings up the matter of data freshness. I know Zacks has some time delay between when an analyst puts out a report and when it gets into their numbers. Consequently when you look at a graph like the one below, the price data is fresh, but the eps data lags by up to a month. That is why when you see a stock move off a base after an eps cut pricewise, you want to see some sort of confirmation from the earnings estimates before jumping in with both feet. Our resident fundamental guru at CA.com (Dick Davis) insists on using Morningstar for this very reason feeling that it has the lowest time lag in the data.
Another contirbuting factor is if we assume the Merrill Lynch analyst is the "good" analyst, the numbers for the rest will be coming down also. With each earnings estimate that is cut, that will put pressure on the stock because people fixate on those cuts especially after the stock has had a very good eps growth rate for quite some time. The hand wringing of "perhaps they screwed up" will start and people will sell.
The lawyers with constant class action lawsuits don't help. (Where were they when Cendant imploded with real criminal actions?)
The time it takes to wash through this sort of earnings estimate water torture varies, but 3 months is my expected dead money time. This might be offset by any IPO hype in the next 6 months or so.
If you look at this graph however, you can see the current drop of the eps estimate on this graph you can see that the eps estimate might not yet be fully disseminated.
marketplayer.com
Stocks rarely spike down in a "V" shape and bounce right back up, although there are exceptions.
marketplayer.com
Or are there other factors in this report that negate the eps downgrade?
Yes, I believe there are, but that would come under the topic of due diligence, not screening. <grin>
The largest factor is the impending spinoff of a division within the next 6 months. That is what started some of this drop in the first place IMO. I believe the company was "clearing the decks" in front of the spinoff and the magnitude of the "clearing" surprised the analysts. The second good thing that was in that report was the 2000 estimate wasn't touched. If there were longer term problems, they would have cut that also. The third good thing in the report was they also left the 5 year growth rate alone implying that nothing has change the longest term outlook of the company.
Disclosure: I own ORB at
$39 (GARP discount of 32%),
$28 (GARP discount of 50%), and
$20 (GARP discount of 65%),
so I get no gold stars for being Mr. Timing which is why I don't do the technical analysis part of the Confirmatory Analysis.com selections, but run some of those numbers out 2-3-5 years even if the current number drops to $1 and see what sort of stock price you get.
marketplayer.com
Worst case I figure is ORB goes back lowered expectations of $1 X p/e of 30-32 which is a stock price of $30-$32 in next few months if nothing else happens. Longer term I don't see why the 30ish multiple cannot be applied to somewhat lowered 2000 estimates so ballpark of mid to upper $40's level is possible. Then if all goes right and the estimates continue to rise past 2000, ...
I don't expect people to use this sort of analysis verbatim, but I believe the concepts are valid.
Does that answer your question?
----
Dave |
